Plant virus identification based on neural networks with evolutionary preprocessing

摘要: In this work, genetic algorithms and multilayer neural networks are applied to plant virus identification. The initial data set is derived via a well known prototype method, which uses specially designed biosensors monitor the reactions. Several techniques have been introduced for preprocessing waves. They include segmentation along time axis fast response, nonlinear normalization emphasize significant information, averaging samples of waves suppress noise effects, reduction in number realize more compact network, etc. Given features acquired time-series signals problem under study, an evolutionary method proposed order produce meta-data from original reduce dimensionality input space, eliminate inherent raw information. A algorithm employed so as smooth out information while, produced sets used training testing producing fitter data. tested against some most commonly classifiers machine learning cross-validation proved its potential towards assisting